The Journal of Environmental Education

ISSN: 0095-8964 (Print) 1940-1892 (Online) Journal homepage: https://www.tandfonline.com/loi/vjee20

The Child in the Garden: An Evaluative Review of

the Benefits of School Gardening

Dorothy Blair

To cite this article: Dorothy Blair (2009) The Child in the Garden: An Evaluative Review of
the Benefits of School Gardening, The Journal of Environmental Education, 40:2, 15-38, DOI:
10.3200/JOEE.40.2.15-38

To link to this article: https://doi.org/10.3200/JOEE.40.2.15-38

Published online: 07 Aug 2010.

Submit your article to this journal

Article views: 8353

View related articles

Citing articles: 71 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=vjee20
 PROGRAM EVALUATION

The Child in the Garden: An

Evaluative Review of the Benefits
of School Gardening
Dorothy Blair

ABSTRACT: Although educators widely use school gardens for experiential education, researchers
have not systematically examined the evaluative literature on school-gardening outcomes. The author
reviewed the U.S. literature on children’s gardening, taking into account potential effects, school-
gardening outcomes, teacher evaluations of gardens as learning tools, and methodological issues.
Quantitative studies showed positive outcomes of school-gardening initiatives in the areas of science
achievement and food behavior, but they did not demonstrate that children’s environmental attitude
or social behavior consistently improve with gardening. Validity and reliability issues reduced general
confidence in these results. Qualitative studies documented a wider scope of desirable outcomes,
including an array of positive social and environmental behaviors. Gardening enthusiasm varies
among teachers, depending on support and horticultural confidence.

KEYWORDS: environmental education, experiential learning, food behavior, school gardens, science
achievement, social behavior

O ver the last 20 years, school gardening has become a national movement. Texas and California
state departments of education and university extension programs have actively encouraged
school gardening by providing curricula and evaluative research (Dirks & Orvis, 2005; Ozer,
2007). Also, 57% of California school principals responding to a statewide questionnaire said that their
schools had instructional gardens or plantings (Graham, Beall, Lussier, McLaughlin, & Zidenberg-Cherr,
2005). Florida, Louisiana, and South Carolina have had programs that promote school gardening (Culin,
2002; Emekauwa, 2004; Smith & Mostenbocker, 2005; University of Florida, 2006).
Northern states have been slower to become involved, but school gardens are no longer exceptional in
cooler climates. In the state of New York, more than 200 schools, 100 teachers, and 11,000 students garden
using a state curriculum (Faddegon, 2005). Vermont actively promotes school gardening in partnership

Dorothy Blair is an assistant professor at Penn State University, College of Health and Human Development.
She is also a member of the Department of Nutritional Sciences. Her teaching and research focus on the inter-
face among food, agriculture, ecology, culture, and social justice. Copyright © 2008 Heldref Publications

WINTER 2009, VOL. 40, NO. 2 15

 with the National Gardening Association, which is housed in Burlington, Vermont (National Gardening
Association, 2006), and provides demonstration gardens, national newsletters, and teacher education.
Overwhelmingly, gardens (Waliczek, Bradley, Lineberger, & Zajicek, 2000) and gardening cur-
ricula target elementary students. Some of the most popular curricula are the 1978 Life Lab K–5
Science Program (LifeLab, 2006); 1990 GrowLab curricula (National Gardening Association, 2006);
Texas A&M’s Junior Master Gardener Program (Dirks & Orvis, 2005); UC Davis’ curriculum
Nutrition to Grown On (California Department of Education, 2005; Morris & Zidenberg-Cherr,
2002); and New York’s curriculum Kids Growing Food (Faddegon, 2005).
School gardening covers a continuum of efforts to increase the horticultural complexity of the
schoolyard, including potted plants, raised beds on asphalt, indoor vermiculture composting, in-ground
plantings (Graham et al., 2005), habitat and butterfly gardens, sunflower houses and ponds, compost-
ing areas accommodating a school’s daily lunch waste (Graham, Feenstra, Evans, & Zidenberg-Cherr,
2004), and a systematic approach to redesign the outdoor space around schools into learning landscapes
(Brink & Yoast, 2004). The purposes of the redesigned schoolyard are academic, behavioral, recre-
ational, social (increased sense of belonging, self-esteem, and compassion), political (the schoolyard as
a visible community asset), and environmental remediation. Educators and landscape architects used
these criteria for the Boston Schoolyard initiative (Corson, 2003) and the Youth and Landscapes pro-
gram, a collaboration between Denver schools and University of Denver graduate students in landscape
architecture to redesign derelict schoolyards (Brink & Yoast).
Schools can move even further afield, as in place-based learning, developing collaborations with
rural community partners that aid and facilitate the study of local natural resources (Emekauwa,
2004), or creating partnerships with university forestry departments, city park naturalists, and
local businesses to facilitate the study of urban forest ecology (Milton, Cleveland, & Bennet-Gates,
1995). Emekauwa reported that 3 years of place-based learning focusing on local ecology—nature
trails, soils, geology, butterfly gardens, and school interactions with community ecological experts—
resulted in substantial reductions in unsatisfactory standardized test scores for language arts, math, sci-
ence, and social studies among fourth-grade students in a poor, rural, 80% African American, Louisiana
school district. Lieberman and Hoody’s (1998) frequently quoted study reviewed 40 schools in 12 states,
comparing classrooms that used the environment as an integrating context for learning with nonintegrat-
ing classrooms. Those researchers found that enthusiasm for learning, standardized test scores, and GPAs
were higher in 92% of the comparisons—particularly in language arts, social studies, science, math, and
thinking skills. The National Environmental Education and Training Foundation (2000) stated that the
environment, “from classroom to schoolyard to local nature centers and parks” (p. 7), enables learning that
is problem-based and interdisciplinary, with a significant positive impact on achievement.
The specific question that I addressed in this review of the literature is whether a school garden, without
causing extensive changes to the schoolyard or integrating broader environmental fieldwork into the cur-
riculum, provides sufficient experiential education to cause measurable and observable changes in student
achievement and behavior. Enthusiasm for school gardening is clearly present, but the literature on school
gardening’s impact on children’s learning and behavior comes from many disciplines and has not yet
received a thorough, integrative review. My approach is to first give an overview of the rationales for school
gardening and then critically examine the evaluative research on school-gardening outcomes.

Rationales for School Gardening

Broadening Children’s Experience of Ecosystem Complexity
In earlier eras, Rousseau, Gandhi, Montessori, and Dewey—most notably—promoted school
gardens (Subramaniam, 2002). When farms and nature were readily accessible to most children, the

16 THE JOURNAL OF ENVIRONMENTAL EDUCATION

goal of school gardens was pragmatic and normative: to teach through experience, to connect chil-
dren to pastoral nature, and to shape their moral outlook (Bundschu-Mooney, 2003; Subramaniam).
School gardening in the United States was originally introduced for aesthetic purposes. It became a
national movement first in 1918 and again, with a focus on food production, during World War II,
but it waned in the 1950s because of the nation’s focus on technology (Subramaniam).
Today’s children lack experience with natural ecosystem complexity. In all, 83% of the U.S. popu-
lation lives in metropolitan areas (U.S. Department of Agriculture, 2006). Thus, pasture or wilder-
ness is no longer the normative standard for experience in nature (Mergen, 2003). Two-worker fami-
lies who are concerned for the safety of their unattended children must choose close supervision of
afterschool and summer playtime. Television, video games, and organized sports have taken the place
of unsupervised wandering and environmental exploration (Moore, 1995). As childhood becomes
more structured, the places where children must play are open and lack the appeal of intimate spaces
grounded in the natural environment (Francis, 1995). City children search out dirt, water, trees, and
natural elements and explore and play in the same manner in which rural children do (Mergen), but
urban sprawl and environmental degradation reduce the frequency of these city children’s positive
experiences with natural elements in their environment (Finch, 2004; Kellert, 2002; Orr, 2002). A
study of three generations of children in a New York City neighborhood shows a decline in natural
areas and an increase in restricted access to the neighborhood and reliance on supervised play (Gaster,
1991). In Gaster’s study, schools were considered safe areas. However, typical asphalt-covered or flat
green schoolyards were, as they are today, monocultures that minimized environmental complexity.
Whether urban or rural, the landscape in which children find themselves is the staging ground for
their imagination, their story, their sense of the world (Mergen, 2003). If formal playgrounds or sports
fields delimit many children’s natural experiences (Nabhan & Trimble, 1994), well-designed school
gardens can readily improve on the complexity of that experience and provide the repetitive access,
meanings, and associations needed to create a bond with a place. However, because of the way school
gardens are typically interpreted and constructed in our culture, few contain intimate spaces, elements
of the wild, or places to dig in dirt. Educators must adjust their norms for neatness, play area supervi-
sion, and ease of outdoor maintenance for school gardens to contain areas that are not neatly planted
or controlled, thereby making them available for children’s imaginative play (Finch, 2004).
Gardens adhering to the principles of biodiversity and organic pest management—containing
ponds or recycling streams, trees, and butterfly attractors—would be havens for a wide variety of
flora and fauna beyond the crops, flowers, and bushes purposely grown and would demonstrate
ecosystem complexity. Gardens that children help to plan allow “close, personal experiences with the
earth” (Thorp & Townsend, 2001, p. 349), repeated sensory contact, and interaction with a par-
ticular intimately known space, creating confidence in the processes of nature that some researchers
believe is necessary for healthy human development (Thorp & Townsend).

Place-Based Learning Clarifies the Nature and Culture Continuum

Personal experience and observation of nature are the building blocks for classroom enrichment
(Nabhan & Trimble, 1994). A garden is an environment in miniature, and to be successful a gar-
dener must work in sympathy with nature (Demas, 1979). Gardens ground children in growth and
decay, predator–prey relations, pollination, carbon cycles, soil morphology, and microbial life: the
simple and the complex simultaneously. Gardens are intensely local. Everything except possibly the
purchased plants and seeds are part of the natural local environment. The clouds, rain, and sun, the
seasonal cycle, the soil and its myriad organisms, the insects, arachnids, birds, reptiles, and mammals
that visit the garden teach about place. Even if some of the weeds, insects, and birds are not native to
a place, these immigrant flora and fauna are as locally adapted as the children themselves. Nature and

WINTER 2009, VOL. 40, NO. 2 17

 natural are relative terms that depend on cultural norms and the limits of our own ahistorical experi-
ence with place (Finch, 2004; Mergen, 2003; Nabhan & Trimble, 1994). Seeds and gardening styles
are the stuff of history, culture, ethnobotany, and literature. Along with English sparrows, starlings,
quack grass, and bees, gardening provides another kind of lesson, one about human interaction with
the natural world.

Vegetable Gardening Teaches Food Systems Ecology

Anonymous prepackaged food arrives at supermarkets from energy-intensive, polluting, and often
obesity-promoting industrial food-manufacturing systems. Researchers have estimated that this system
consumes 17–20% of American fossil fuel and that 29% of the food is wasted (Blair & Sobal, 2006;
Pollan, 2006). To decrease the threat of the obesity epidemic, children need to broaden their perspective
on what foods are edible and to repersonalize food. Gardening in America’s northern regions during the
school year requires elongating the growing seasons in both spring and fall, thus stretching children’s
knowledge and taste for cool-season vegetables, particularly for dark leafy greens. Because of our super-
markets’ global reach and constant supply of heat-loving vegetables, many cool-season crops remain
unfamiliar. For more ecological, local food systems to satisfy year-round vegetable needs, children’s
tastes in food need to expand beyond the fatty, salty, sweet, and subtropical (Blair, 1996).
School and youth gardens teach “how a plant goes from seed to plate” (Rahm, 2002, p. 175), as
one master gardener said. Such gardens introduce young gardeners to local sustainable food systems,
as children eat their own produce, compost cafeteria food waste, and connect with adult growers and
market gardeners (Graham et al., 2004; Moore, 1995; Morris, Briggs, & Zidenberg-Cherr, 2000).
The act of growing food from seeds is exciting, even miraculous; the product is something special
to be taken home to share. This sentiment is expressed by Thorp and Townsend (2001) in the fol-
lowing statement:

[G]ardening changes the status of food for all involved. When one gardens, food can no longer
be viewed as a mere commodity for consumption; we are brought into the ritual of communal
goodness that is found at the intersection of people and plants. Food that we grow with our own
hands becomes a portal for personal transformation. (p. 357)

Exposure to Nature and Gardening in Childhood Shapes Adult Attitudes and Environmental Values
Many authors and researchers believe that today’s children lack the exposure to the natural world
that shapes environmental values and puts science in context (Bundschu-Mooney, 2003; Finch,
2004; Kahn, 2002; Kellert, 2002; Orr, 2002). Chawla’s (1998) review of the qualitative and sur-
vey literature found that adults who had significant and positive exposure to nature as children—
experiences, often with significant adults, that socialize them to view nature in positive and meaningful
ways—were more likely to be environmentally sensitive, concerned, and active. In a sample of teenage
natural-resource workers, Vaske and Kobrin (2001) showed that a teenager’s identity with a place medi-
ated the relation between dependency on the place and environmentally responsible behaviors.
Active childhood involvement with plants may affect subsequent attitudes and behavior in adults. Blair,
Giesecke, and Sherman (1991) found that minority participants—African Americans from the South and
Asian immigrants—in community gardening projects in Philadelphia had gardened as children in rural
areas. Lohr and Person-Mims (2005) studied metropolitan adults’ attitudes toward trees and gardening in
relation to their memories of their home surroundings in childhood, their time spent in outdoor places,
and their time spent actually performing gardening activities (telephone survey of 112 most populated
U.S. cities; response rate = 52%; sample size = 2,004). Active gardening in childhood was the most impor-
tant predictor of whether trees had personal value in adulthood.

18 THE JOURNAL OF ENVIRONMENTAL EDUCATION

 Francis (1995) obtained qualitative interviews from 100 Californian and Norwegian gardeners
that explored the significance and meaning of places that they described in their childhood memories
of gardens. All of the respondents had vivid memories of favorite private places in gardens that were
protected, sheltered, or hidden. Francis wrote:

Our interviews suggest that garden meaning is a complex ecology of idea, place and action. We
found that when children become involved as gardeners or farmers rather than as passive observers
of gardens, a deeper significance and meaning is established. Gardens that operated on all levels
simultaneously—as idea, place and activity—can become sacred places. (p. 8)

School Gardening: A Broader Effect Than Experiential Education?

The style of learning that happens in school gardens, using direct contact with natural phenomena,
is considered experiential, inquiry-based learning grounded in concrete experience (Corson, 2003;
Kellert, 2002; Mabie & Baker, 1996; Rahm, 2002). Kellert argued that because nature changes
rapidly, it attracts and stimulates a child’s attention. Naming and categorizing objects found in the
particularly information-rich and potentially fascinating natural world facilitates children’s capacity
to retain information and ideas, a first step in cognitive development, as Bloom’s taxonomy of cogni-
tion outlined (Bloom, Engelhart, Furst, Hill, & Krathwohl, 1956). The experience of nature’s detail,
whether direct or vicarious, provokes children’s need to comprehend and make sense of what they
have experienced. According to Kellert, it is “a rich diet for cognitive development” (p. 125), giving
ample opportunity for creative inquiry; “strengthening the cognitive muscle we call mind[;] and
developing and reinforcing the child’s capacities for empirical observation, analytical examination,
and evidentiary demonstration” (p. 125).
Although higher order cognitive skills are useful in many areas of life, in schools they are most
often a focus of the math and science curricular reform. An overemphasis on factual knowledge
has led to weakness in processing skills and critical thinking in the average U.S. student (Culin,
2002; Gibbs & Fox, 1999). South Carolina’s statewide school program of butterfly-garden train-
ing addressed these science-education deficits through experiential learning that integrated science,
math, writing, and social studies. Children stocked the gardens with flowers grown from seed, kept
census counts of 27 butterfly and caterpillar populations, tagged migrating monarchs, and interacted
with students in Mexican schools (Culin).
Two studies examining the outcomes of experiential learning have shown that it effectively stimu-
lated higher orders of cognition; one of these studies showed that gardening was no more effective than
other hands-on agricultural projects as a stimulator. Waliczek, Logan, and Zajicek (2003) evaluated
the impact of a 4-hr outdoor hands-on nature program regarding weather, insects, water, and soil on
the critical thinking and cognition of 175 second- to fourth-grade students from five New Mexico
schools. Researchers interviewed students, teachers, and volunteers about the outcomes by using open-
ended questions, and then they classified keywords and phrases as one of the six categories of Bloom’s
taxonomy of cognition. In all, 87% of respondents used application terminology, 19% used analysis
terminology, and 26% used evaluation and synthesis terms such as problem solving, integrate, plan, test,
and support. Mabie and Baker (1996) assessed the impact of two 10-week experiential educational
interventions—(a) a school garden project and (b) three discrete in-class projects of seed starting, chick
rearing, and bread baking—on the science-processing skills of 147 Hispanic and African American Los
Angeles middle school students in comparison with a control science class taught as usual. Qualitative
preassessments and postassessments consisted of written and verbal responses to a series of unrelated
hands-on cognitive tests. Both treatment groups improved their posttest scores dramatically, show-
ing increases in observational, ordering, comparison, and communication science-processing skills,

WINTER 2009, VOL. 40, NO. 2 19

 with neither improvement in the control group nor difference by treatment. From these few studies,
researchers can reasonably conclude that experiential learning, rather than gardening per se, improves a
child’s chances to use higher order cognitive skills.
However, these two types of experiential learning may not be equal in other ways. The difference
between a structured discrete experiential learning experience and a long-term involvement in a
gardening process resides in the multitude of unstructured learning opportunities that are not in the
lesson plan, happen spontaneously and nonhierarchically, and involve students and their adult men-
tors in multidirectional learning (Milton, Cleveland, & Bennet-Gates, 1995; Rahm, 2002; Thorp
& Townsend, 2001). Gardening requires physical labor. Repetitive tasks give ample opportunity
for informality, and results happen slowly over a long time. Rahm studied the conversations that 6
African American 11–14-year-old students held with master gardeners and with each other during
an 8-week summer youth-gardening project in the Midwest, where young interns prepared the soil,
started seeds, nurtured, harvested, and marketed their produce. The project embedded informal
science education in gardening conversation that flowed in a natural and organic way, involving,
as Rahm stated, “sense making through discourse” (p. 179). As Rahm also stated, “Youth were the
creators and not merely the consumers of the science curriculum” (p. 180). In addition, as Thorp and
Townsend noted, “[A]t the heart of scientific inquiry is good old-fashioned slack-jaw wonder. ‘Mrs.
Thorp look how big this turnip is!’” (p. 356).
The very qualities that render school gardening a potent and multidimensional experiential-
learning experience—being outdoors and involved in hands-in-dirt digging, planting, and cleanup—
may render it unpopular with teachers who prefer the safety, predictability, cleanliness, and ease of
the indoor classroom.

School-Gardening Studies That Have Assessed Learning Outcomes

Using a questionnaire mailed to 17 school-gardening researchers (76% response rate), Phibbs and
Relf (2005) found that the learning outcomes most often studied were health and nutrition (69%),
environmental education (EE; 30%), and self-esteem or self-concept (30%). The age groups studied
were predominantly elementary (85%) or middle school (38%). The present research also shows
that among published quantitative studies, science achievement, nutrition knowledge, and change
in food behavior have been most frequently measured, preceding environmental attitude change,
self-esteem, and life skills. The intended research subject of qualitative studies is most frequently
agricultural education, but the results are much wider in scope.

Quantitative Assessment of the Learning Impact of School Gardening

In all, 12 studies have used quasi-experimental pretest and posttest designs or simple posttest
designs to quantify the impact of school-garden participation on children’s learning or behavior. The
researchers tended to study third- to sixth-grade students. To a lesser degree, the researchers studied
general elementary school students; they studied first-grade students only once. Tables 1–4 sum-
marize these research reports and categorize them by specific learning outcome for gardening and
nongardening students.
Using a criteria of p < .05 for significant results, 9 of the 12 studies revealed a positive difference
in test measures between gardening students and nongardening students. School gardening increased
the science scores in all reported studies. Gardening improved elementary student preference for
vegetables as snacks in Texas (Lineberger & Zajicek, 2000) and California (Morris & Zidenberg-
Cherr, 2002). A 12-week gardening intervention increased fruit and vegetable consumption among
sixth-grade students in Idaho. However, gardening was no more effective than simple nutrition les-
sons in conveying nutrition knowledge (Morris & Zidenberg-Cherr). With gardeners serving as their

20 THE JOURNAL OF ENVIRONMENTAL EDUCATION

own control, Skelly and Zajicek (1998) reported that the degree of improvement in environmental
attitude regarding their gardening subjects was directly related to the amount of outdoor activities
experienced in the garden. However, few studies examined the environmental and behavioral impacts
of gardening, and those studies’ results were mixed.
Out of those studies, three did not meet my criteria of significance, although only one found no
significant difference. Mabie and Baker (1996) provided no statistical analysis but reported that garden-
ing groups had higher agricultural achievement scores than did students doing in-class projects. Using
significance criteria of p < .10, Waliczek and Zajicek (1999) found a small difference in environmental
attitude change in relation to school gardening in a diverse sample of 598 Texas and Kansas second- to
eighth-grade children. However, the pretest–posttest difference was less than 1%. Given the size of the
sample, it is possible that the null hypothesis should have been accepted. A second report using the same
sample examined gardening’s effect on self-esteem and life skills and showed no significant difference
(Waliczek, Bradley, & Zajicek, 2001). Although the participating schools differed in how they handled
the gardening experience, Waliczek et al. reported no oversight or monitoring criteria. Those authors
mentioned a significant interaction between the variables of school and attitude toward school. On the
basis of post hoc comparisons, those authors hypothesized that students had better attitudes toward
schools that allowed more individual participation in the garden.
The quasi-experimental designs of the studies in Tables 1–4 have left them open to criticism and
the probability of false positives. Rather than randomly assigning classes to a treatment method, in
most cases the researchers allowed teachers to volunteer their classrooms for the experimental group.
Teachers differ significantly in their training and enthusiasm for gardening (Graham et al., 2005).
Enthusiastic teachers are far more likely to volunteer their classrooms and promote positive experi-
ences, biasing a study’s results. Dirks and Orvis (2005) reported significant classroom effects in one
study of science achievement. They also found ethnicity effects. White students’ environmental atti-
tude increased with gardening more than the attitude of African American students did (Waliczek &
Zajicek, 1999). The ethnicity of control and experimental groups was inadequately matched in some
of these studies. I address methodological issues for improvement of quasi-experimental designs and
methods in more depth in the Discussion section of this article.

Qualitative Studies of School Garden Effects

The purpose, research questions, and results of seven qualitative studies of elementary school-
gardening projects are presented in Table 5 (triangulated data) and Table 6 (case studies). Methods
and research questions vary. The results of several of these studies were discussed earlier in the pres-
ent article.
Whatever the original intent of these qualitative studies of elementary school gardening, common
themes run through each report and are listed below.

1. All seven studies reported that students were delighted and highly motivated by the pleasures
of gardening and the opportunity to get dirty outside and were excited by exploratory learning
framed in a garden context (Alexander, North, & Hendren, 1995; Brunotts, 1998; Brynjegard,
2001; Canaris, 1995; Faddegon, 2005; Moore, 1995; Thorp & Townsend, 2001).
2. All seven studies reported that students showed improved school attitude and pride in the
garden and its produce. The students involved their parents, who became more involved with
school. (Alexander et al., 1995; Brunotts, 1998; Brynjegard, 2001; Canaris, 1995; Faddegon,
2005; Moore, 1995; Thorp & Townsend, 2001).
3. All seven studies reported that school gardens had a strong community-building component,
promoting teamwork, student bonding, a broader range of interaction with adults, and community

WINTER 2009, VOL. 40, NO. 2 21

22
TABLE 1. Quantitative Assessment of Science Achievement in Conjunction With School Gardens

Authors, by date
published Objective Sample or design Tools Results

R. Mabie and Assess 2 types of activities Hispanic and African Instrument developed by No statistical analysis
M. Baker (1996) to improve student American 5th–6th-grade researchers and tested in a reported
knowledge of Ag students from 2 inner-city Los Angeles school; Kuder- Gardening groups showed
Los Angeles schools; Richardson 20 reliability = .74 the biggest increase in
control group (n = 31), Gardening group: gardened interest and knowledge of
gardening group (n = 56), 1 hr per week for 10 weeks Ag
Ag project group (n = 57) Project group: bread baking, Project group was
Pre- and posttest chick rearing, and seed intermediate in pre- and
germination for 3 days posttest change between
control and gardening
group
C. D. Klemmer, Develop cognitive test 3rd–5th-grade students from Test comprising 40 multiple- Cronbach’s α = .86 overall;
T. M. Waliczek, instrument to assess SA 7 central Texas schools choice questions based on .92 for 5th-grade students
and J. M. gains in conjunction with using state youth- Bloom’s Taxonomy of
Zajicek (2005a) gardening gardening curriculum Education Objectives
(N = 657)
C. D. Klemmer, Assess effectiveness of school Same sample as above (n = Texas state youth-gardening SA higher (p < .001) for
T. M. Waliczek, gardens in enhancing SA 453 in experimental classes; curriculum given to gardening students
and J. M. n =194 in control classes experimental groups Only 5th-grade students
Zajicek (2005b) mixed in schools) delayed for control had sig. dif. (p < .001)
Posttest only Klemmer SA test

THE JOURNAL OF ENVIRONMENTAL EDUCATION

 L. L. Smith and Quantify effects of school 5th-grade students of African Used Jr. Master Gardener Higher SA in gardening
C. E. gardens on SA of 5th-grade American or multiple Handbook horticulture students (p < .05)
Mostenbocker students ethnicities from 3 inner-city undergrads gardened with
(2005) Baton Rouge schools; students once per week for
control (n = 57) and 14 weeks
experimental (n = 62) Klemmer SA test
classes in each school
Pre- and posttest
A. E. Dirks and Evaluate the impact of Jr. 277 students from 11 Developed testing tool Small but sig. dif. (p < .0001)

WINTER 2009, VOL. 40, NO. 2

K. Orvis Master Gardener Program Indiana schools with few measuring short-term in K&A
(2005) on classroom SA minorities K&A change Significant classroom effect
Pre- and posttest on K&A and school-
garden effect on attitude
(p < .0001)

Note. SA = science achievement; Ag = California food and fiber industry; K&A = knowledge and attitude; sig. dif. = significant difference.

23
24
TABLE 2. Quantitative Assessment of School Garden Food and Nutrition Outcomes

Authors, by date
published Objective Sample or design Tools Results

S. E. Lineberger Test garden-activity guide
and J. M. for teachers
Zajicek (2000) Measure attitude change
toward F&V
J. L. Morris, A. Assess feasibility of school-
Neustadter, gardening program with
and S. Zidenberg- 1st-grade students for nutr
Cherr (2001) education and food-
behavior change
J. L. Morris and Develop and evaluate 9-lesson
S. Zidenberg- garden-enhanced nutr-
Cherr (2002) education curriculum
111 Texas 3rd–5th-grade
students from 5 schools
(quasi-experimental)
No control group
Pre- and posttest
1st-grade students from 2
California schools matched
for location and ethnicity,
with (n = 48) and without
(n = 49) school gardens
(quasi-experimental)
Pre- and posttest
213 California 4th-grade
students (control school,
NL school, NG school)
Teachers guide with 34
gardening activities
F&V preference
questionnaire
24-recall food journals
(pre and post)
Nutr knowledge score based
on food-group recognition
Willingness to taste 6 veg,
preference, and ability to
name
Nutr knowledge questionnaire
Willingness to taste 6 veg:
carrots, broccoli, spinach,
Increased veg preference
(p < .05)
Fruit preference had no
change
Increased preference for
F&V as snack (p < .01)
No change in 24-hr recall
pre- and posttest
Gardeners had a higher
score on food-group
identification (p < .02)
Gardeners more willing to
taste veg (p < .005)
No preference or naming
differences
No difference in nutr
knowledge between NL
and NG sites

THE JOURNAL OF ENVIRONMENTAL EDUCATION

Pre- and posttest and follow-up snow peas, zucchini, and No difference in willingness

WINTER 2009, VOL. 40, NO. 2
J. D. McAleese Assess effects of garden-based 122 Idaho 6th-grade students
and L. L. nutr education on F&V from 3 elementary schools;
Rankin (2007) consumption 2 control schools without
gardens and NL school
randomly assigned; NG
school was a convenience
choice because of walking
access to a raised-bed
garden (nonrandomized)
Pre- and posttest
Note. Nutr = nutrition; F&V = fruit and vegetable; veg = vegetables; NL = nutrition lessons only; NG = nutrition lessons plus gardening.
25
jicama; rated veg for
preference on Likert-type
scale
6-month follow-up on
preference
99 students from 3 schools
completed 3 pre- and post-
24 hr food recalls
NL and NG schools’
intervention: Nutrition in
the Garden (12-week
horticulture nutr
curriculum)
NG school also had 12-week
hands-on garden-based
activities
to taste veg among sites
Compared to others, NG
site preferred more veg:
snow peas (p < .05) and
zucchini (p < .0005)
6-month follow-up: NG
group retained significantly
higher preference for
broccoli, zucchini, and
snow peas
F&V consumption changed
only at NG school where
cafeteria servings of both
more than doubled (p <
.001).
Vitamins C and A and fiber
intake increased at NG
school (p < .016–.001)
26
TABLE 3. Environmental Attitude Change in Conjunction With School Gardens

Authors by date
published Objective Sample or design Tools Results

S. M. Skelly and Develop an interdisciplinary Sample is 2nd- and 4th-grade Project GREEN: developed Sig. dif. in Environmental
J. M. Zajiek approach to environmental students by class from 4 33 activities for school Response Inventory (p <
(1998) education using gardens Texas elementary schools; gardens .001); dose response =
Test treatment difference in experimental group (n = 153), Children’s Environmental score increase with
EA control group (n = 84) Response Inventory number of outdoor
Posttest only activities
Negative age effect (p < .05)
T. M. Waliczek Measure changes in EA with 589 2nd–8th-grade students Project GREEN curriculum Pretest score was 31.45;
and J. M. Project GREEN gardening from 7 elementary schools Author-created EA scale posttest score was 31.71,
Zajicek (1999) activities in Texas and Kansas t = –1.712 (p < .10)
Pre- and posttest design Significant effect of gender
and ethnicity

Note. EA = environmental attitude; sig. dif. = significant difference.

THE JOURNAL OF ENVIRONMENTAL EDUCATION

 TABLE 4. Assessment of School Gardening’s Effect on Self-Esteem and Life Skills

Authors, by date

WINTER 2009, VOL. 40, NO. 2

published Objective Sample or design Tools Results

T. M. Waliczek, To determine if Project
R. D. Bradley, GREEN curricula
and J. M. positively influenced
Zajicek (2001) students’ interpersonal
relationships and attitudes
C. W. Robinson Assess changes in life-skill
and J. M. development of elementary
Zajicek (2005) students participating in a
1-year gardening program
598 2nd–8th-grade students
in 7 Texas and Kansas
schools; gardening
children compared with
“norming” population for
testing instrument
Pre- and posttest design
281 3rd-, 4th-, and 5th-grade
students from 7 Texas
elementary schools
Pre- and posttest with
experimental classes (n =
190) and control classes
(n = 91)
Project GREEN activities
for school gardens was used
for 6 months (not clear
how well teachers carried
out gardening activities)
The Self Report of Personality
Scale for children and
adolescents
Texas Extension Service
school-gardening
curriculum with teacher
training
Youth Life Skills Inventory
with 3-point Likert-type
scale
No sig. dif. between
gardening children and
“norming” control
No significant effect of
gardening on experimental
group’s attitudes or
relationships
Small but significant
increase (p < .05) in scores
of gardening children
No change in controls but
gains in areas of working
with groups and self-
understanding

Note. Sig. dif. = significant difference.

27
28
TABLE 5. Triangulated Studies of Grade-School Gardens Using Qualitative Methods and Analysis

Authors, by date
published Objectives or questions Sample or research design Observations or themes

J. Alexander, Pilot study identifying and evaluating Participating inner-city San Antonio
M.-W. North, short-term effects of master gardener School 2nd- and 3rd-grade students;
and D. K. classroom-gardening project 3 gardening classes, 2 nongardening
Hendren (1995) controls
Observations and interviews with
principal, 5 teachers, master gardener,
52 students, and 3 parents
C. M. Brunotts Evaluation of Pittsburgh Civic Gardening Socioeconomically depressed school
(1998) Center’s 14-session school-gardening 150 Kindergarten and 2nd- and 4th-
outreach program grade students in groups
Teachers (open-ended questions); 13%
of parents (open- and closed-ended)
Methodology was a post hoc adaptation
to various problems with pre- and
postdesign
Moral development related to valuable
life lessons embedded in gardening
Academic learning related to hands-on
nature experimentation
Gardens engendered parental support,
enthusiasm, and involvement
Student pleasure, self-satisfaction,
teamwork
Master gardener critical element as role
model
Fear of vandalism, protective feelings
Gardening viewed as valuable and
effective augmentation of science
curriculum, broadening student
horizons, and increased parental
involvement; gardening increases
student excitement about learning,
pride, hope, fun, ability to work
cooperatively, and taking
responsibility; increased
observation skills and
environmental caretaking

THE JOURNAL OF ENVIRONMENTAL EDUCATION

Teacher training needed
 L. Thorp and Phenomenological understanding of the
C. Townsend impact of a garden-based agricultural
(2001) education curriculum on teachers and
students in a low-income, multiethnic,
midwestern elementary school, K–5
Does gardening improve agency and
connection of students to school?
Which lessons facilitate learning?
Which lessons have barriers or
constraints?
WINTER 2009, VOL. 40, NO. 2
P. A. Faddegan Evaluate New York’s Kids Growing Food
(2005) minigrant curricular effects
Perceived increase in agricultural literacy;
identify other outcomes
29
Purposive sampling of 5 teachers and 40
students
Multiple qualitative methods: interview,
conversation, document analysis, and
photo elicitation
Naturalistic data analysis, refining
questions over time
Content analysis of final reports, open-
ended survey questions with teachers,
focus groups with school gardeners,
personal interviews (Population >
11,000 students; > 200 gardens)
Garden reshaped school culture, creating
hope, growth, and community
Garden rhythms improved sense of
control and place
Children experienced “comfort, security,
belonging, pleasure and wonder”
(p. 357)
Garden was a place of self-expression
and feeling uniquely present
Food took on new, deeper meanings
Requires dedicated volunteer
Most teachers did not fully understand
the term agricultural literacy and
needed more training
Gardening’s greatest success was
improving attitudes toward school,
motivation, and enthusiasm in all
related subject areas, in particular
plants, ecology, and nutrition
Garden provided context for experiential
learning for core curricular subjects
30
TABLE 6. Case Studies of School-Gardening Projects and Outcomes (Direct Involvement of Author)
Author, by date published Place or purpose of garden
I. Canaris (1995) Vermont 1st–4th-grade mixed class worked with organic
farmer to develop snack garden and learn where food
comes from (2-year teacher retrospective)
R. Moore (1995) Berkeley, CA, primary school environmental-schoolyard
design project retrospective, a biodesign developed from
blacktop over a 10-year period with collaboration from
university faculty, students, and an outdoor resource teacher
Purposes were to develop an environmental pedagogy where
all subjects would be taught through the outdoors and to
promote sustainable development values
S. Brynjegard (2001) Napa, CA, elementary-school garden established by author as
AmeriCorps volunteer to promote environmental awareness
Interviews with students, parents, and teachers at 3 gardening
elementary schools in the San Francisco Bay area
Research question: Do children gain unique insights into
environmental issues through gardening?
THE JOURNAL OF ENVIRONMENTAL EDUCATION
Gardening impacts
Increase in nutrition, food literacy, gardening and math skills,
observation and connection to nature
Gardening connected the students to the community,
stimulating student creativity, recycling, fundraising, and
parental involvement
“Gardens provided the most direct source of children’s
emotional involvement with living systems. They
accommodated every stage of the learning cycle, stimulated
by a diversity of flowers and vegetables, constantly
changing, interacting with their surroundings, adapting to
new circumstances, as children counted, measured,
observed, described, interpreted and recorded.” (p. 75)
“Garden projects had the unique capacity to generate a
collective sense of purpose through the shared experience
of getting one’s hands in the soil. No other activity
duplicated such an intimate combination of freedom of
expression and discipline.” (p. 79)
Gardening fosters detailed nature observation leading to
mind and heart understanding of natural interactions,
identification with and appreciation for living things
Gardens need coordinator, whole school support, and unity
Children’s responsibility and decision making promotes
attachment, empowerment, and ownership
outreach (Alexander et al., 1995; Brunotts, 1998; Brynjegard, 2001; Canaris, 1995; Faddegon, 2005;
Moore, 1995; Thorp & Townsend, 2001).
4. All seven studies found that school gardens provided a diversity of environmental-
stewardship, math, and science-education opportunities: measuring space, observing and
experimenting with natural and plant processes, learning about soil improvement, recycling,
creatively reusing materials, propagating, germinating, and saving seeds (Alexander et al., 1995;
Brunotts, 1998; Brynjegard, 2001; Canaris, 1995; Faddegon, 2005; Moore, 1995; Thorp &
Townsend, 2001).
5. Out of the seven studies, four described how vegetable gardens provided holistic food and
nutrition education, food-systems thinking, tasting, snacking, cooking dinners, food sales and
philanthropy, and good food as reward for good work (Canaris, 1995; Faddegon, 2005; Moore,
1995; Thorp & Townsend, 2001).
6. In addition, four of the seven studies reported that a nonstructured, discovery approach
successfully provided the students with opportunities to explore natural phenomena (Brynjegard,
2001; Canaris, 1995; Moore, 1995; Thorp & Townsend, 2001).
7. Four of the seven studies emphasized that school gardens required dedicated, experienced
adult volunteers, master gardeners, or paid coordinators to flourish over time (Alexander et al.,
1995; Brynjegard, 2001; Canaris, 1995; Thorp & Townsend, 2001).
8. Last, two of the seven studies noted that many elementary teachers were not agriculture-
literate and lacked knowledge of basic plant science or plant-growing skills (Brunotts, 1998;
Faddegon, 2005).

Four of the seven qualitative studies of K–12 gardening were evaluations of gardening projects
that the authors had initiated or were directly involved in. Those studies opened themselves to the
danger of overly enthusiastic reporting and biased analysis. However, those authors were also in the
best position to unravel the garden–child interactions. Researchers should understand and evaluate
such studies as best-case scenarios.
Studies of gardening involving high school students as participants are rare. A review by Sullivan
(1999) briefly mentioned a project at a rural health center in Arizona where local high school stu-
dents and project staff tended a demonstration and community garden next to the health center to
provide technical support and encouragement for home gardening in the local area. Then the same
high school students provided technical expertise for these new home gardens. Horticultural therapy
has been successfully used to increase self-confidence, pride, and self-esteem among troubled youths
in Ohio (Hudkins, 1995). However, I found no quantitative and only two qualitative studies con-
necting gardening with high school students. Although they did not fit the pattern of in-school gar-
dening encountered by researchers in elementary schools, those studies showed innovative ways for
using gardening with older students. Lekies, Eames-Sheavly, Wong, and Ceccarini (2006) reported
on a New York State 4-H children’s garden consultant program in which 7 girls served as consultants
to adults in the design of children’s school gardens. Those researchers described the process of men-
toring those girls through activities and garden site visits to the point where they were competent
to assist the adults. The researchers concluded that the mentored girls gained empowerment and
self-esteem and provided valuable improvements to the children’s garden-design site and garden
programming. Krasny and Doyle (2002) conducted qualitative, triangulated research on Cornell
University’s six-city garden mosaics program, engaging inner-city youth attending summer pro-
grams in participatory research with adult gardeners in their communities. Their research included
interviews with 4 gardeners, 11 community educators, and 28 predominantly African American
and Hispanic 9–16-year-old participants. Youth enhanced their gardening, teamwork, and research

WINTER 2009, VOL. 40, NO. 2 31

 skills. They formed learning, helping, and often personal relationships with the gardeners with whom
they studied. These studies showed ways to successfully engage high school students in gardening,
improving their skills and self-esteem.

Studies of Principals’ and Teachers’ Evaluations of the Effectiveness of School Gardens

Another approach to evaluating school-garden effectiveness is for researchers to measure (a) princi-
pals’ and teachers’ enthusiasm for gardening as a learning tool, (b) how teachers find gardens useful,
and (c) what barriers they perceive in the integration of gardens into the curriculum. Studies of prin-
cipals or teachers have involved: (a) sending questionnaires to schools or teachers whom researchers
identified as garden users, (b) sampling all schools in an area, or (c) identifying schools that have
gardens and interviewing teachers in those schools. Approaches (b) and (c) obtain the views of users
and nonusers and are reviewed separately.
DeMarco, Relf, and McDaniel (1999) sent a national school-gardening survey to 322 elemen-
tary school recipients of a National Gardening Association Youth Gardening Grant. The usable-
survey rate was 73%. The researchers did not specify who responded to the questionnaire, but they
implied that participating teachers were the respondents and could be potentially biased as grant
recipients. Less than 5% of respondents felt that school gardening was unsuccessful at enhancing
student learning, and 61% felt that it was very successful at enhancing student learning. Goals
for school gardening were academic (92%), social development (83%), recreational (63%), and
therapeutic (52%). The subject areas that at least 50% of teachers reported that they taught in con-
junction with the school garden were science (92%), EE (83%), mathematics (69%), language arts
(68%), health and nutrition (59%), ethics (58%), and social studies and history (51%). DeMarco
et al. used cluster analysis to identify those factors most frequently indicated as essential for school-
gardening success. These include a person responsible for school-gardening activities, site and materi-
als availability, and support from the principal. Participants selected the factors of student ownership
and integration with other subjects as crucial for school-garden success.
In a similar study of 35 schools and 71 Florida elementary teachers who had entered their gardens
in a 1997 University of Florida contest (100% response rate), Skelly and Bradley (2000) found that
teachers used gardens for EE (97%), to help students learn better (84%), for experiential learning
(73%), and because the teacher had a personal love of gardening (67%). Most teachers were encour-
aged by their administration (54%). Also, 85% of students spent between 1 hr (68%) and 2–3 hr
(17%) per week in the garden, but usually they spent more time on gardening subjects in the class-
room (Skelly & Bradley).
California researchers provided three studies of attitudes and perceptions about school garden-
ing in schools where gardens of some type existed but where respondents were not necessarily
enthusiastically involved in gardening. Graham et al. (2005) sent a questionnaire by e-mail and by
nonelectronic postal service to all California principals (43% response rate). Of those responders,
57% (2,381) indicated that their schools had some kind of garden that grew flowers or vegetables.
The gardens ranged from in-ground gardens (69%), to raised-beds gardens (60%), to gardens in pots
(46%). Those various forms of gardening opportunities were used primarily for academic instruction
in kindergarten to eighth-grade science (86%), environmental studies (64%), nutrition (63%), lan-
guage arts (62%), and math (58%); and in high school science (74%), environmental studies (54%),
nutrition (40%), and agricultural studies (42%). Principals thought that gardens were moderately
to very effective in enhancing science education (69%). The factors that most limited combining
classroom instruction with gardening were (a) lack of time, funding, staff support, and curricular
materials linked to academic standards; and (b) lack of teacher knowledge, training, experience, and
interest in gardening.

32 THE JOURNAL OF ENVIRONMENTAL EDUCATION

 Graham and Zidenberg-Cherr (2005) used a mailed questionnaire to survey a subset of the fourth-
grade teachers at schools that the previous questionnaire indicated had gardens. The response rate
was 36% (n = 592). In all, 68% of responders used gardens for instruction. Echoing their principals,
these fourth-grade teachers thought that gardens were most important for enhancing academic
instruction (72%), but they also thought that gardens effectively enhanced social skills (41%). The
percentage of teachers indicating that gardens were moderately to very effective at enhancing specific
skills and subjects ranged from 53% for science to only 25% for healthy eating habits. Notable were
the number of no opinion answers for every category (26–40%) and the high percentages of teach-
ers who felt that gardens were not effective, slightly effective, or only somewhat effective. However,
some of the gardens at these schools were minimal, and many grew no food. Like the principals, the
teachers indicated that lack of time (67%), lack of teacher interest (63%), lack of experience (61%),
and lack of knowledge (60%) were major barriers to using gardening for instruction.
Graham et al. (2004) obtained a 59% response rate to a distributed questionnaire aimed at all
teachers (N = 118) participating in Farm to School Connections, a three-school pilot program in
Davis, California, that combined school gardens, cafeterias, and classrooms to improve elemen-
tary school children’s eating habits. These three schools had a garden coordinator (a retired and
experienced teacher paid by the state), multiple gardens, farm-to-school salad bars, and school
lunch-plate-waste composting projects. These teachers were more receptive to school gardening
than were the general sample of California fourth-grade teachers. The percentages of teachers using
gardens to teach academic subjects were 90% for science, 71% for nutrition, 64% for language arts,
60% for environmental studies, 59% for health, 57% for agricultural studies, and 56% for math.
Perceived barriers were lack of time and lack of curriculum linked to standards, but cited barriers did
not include lack of teacher interest, training, or knowledge of gardening.
A major methodological problem with the aforementioned mail-and-distribution studies was non-
response bias. Response rates are higher when respondents have a special interest or involvement in
the topic of study (Donald, 1960). DeMarco et al.’s (1999) and Skelly and Bradley’s (2000) studies of
gardening teachers had response rates of 73% and 100%, respectively. Graham et al.’s (2004) study of
teachers adequately supported at gardening schools had a 59% response rate. Teachers from schools
where gardens were promoted at the state level without perceived adequate attention to time and
training had a 36% response rate (Graham & Zidenberg-Cherr, 2005). Unmotivated nonresponders
were likely to differ systematically from the responders in variables most critical to interpreting the
study results (Ellis, Endo, & Armer, 1970). In the case of those mail-and-distribution studies, teacher
and administrator attitudes toward the efficacy of school gardening in terms of learning outcomes
was the most critical variable.

Discussion
Research Question and Methodological Issues
The question addressed by this review of the literature is whether a school garden, without educa-
tors’ either changing the schoolyard extensively or integrating broader environmental fieldwork into
the curriculum, would provide sufficient experiential education to cause measurable and observable
changes in student achievement and behavior. The results of the reviewed research were positive. In
all, 9 out of 12 quantitative studies reinforced the results of Lieberman and Hoody (1998), showing
increased science achievement and behavioral improvement in schools that use school gardening
as their integrating context for learning. Also, 9 qualitative studies unanimously reported positive
learning and behavior effects of school gardening or garden involvement. Mabie and Baker (1996),
Rahm (2002), and Waliczek et al. (2003) have shown a positive impact of outdoor gardening or

WINTER 2009, VOL. 40, NO. 2 33

 nature programs on higher order cognitive skills. Teacher surveys showed that academic achievement,
particularly in the area of science, was the most frequently cited reason for using school gardens.
Gardening studies are most commonly performed with third- to sixth-grade students, although
researchers cannot say that younger or older children would not benefit.
However, given the methodological problems mentioned previously, the research hypotheses
addressed by the qualitative studies cannot be uniformly affirmed. Short-term, quasi-experimental
designs are not considered valid or reliable. Systematic biases in data-collection techniques imply that
the results reported by the quantitative researchers were most likely more positive than was valid. The
quantitative studies I have reviewed suffered from lack of both rigorous sampling procedures and
random assignments of control and experimental groups. At least four studies used testing instru-
ments without proven validity.
In designing their studies, future researchers will need to control the previously uncontrolled
teacher and classroom effects, ethnicity effects, and nonresponse biases. Teacher attitudes toward
gardening, EE, and experiential education should be control variables in any quantitative study
where random selection of gardening classrooms is not an option. Use of propensity scores is a way of
accounting for the bias inherent in nonrandom assignment to groups, as in these quasi-experimental
studies of the effects of gardening on student achievement, attitude, and behavior. Researchers can
evaluate teachers and schools on important variables, such as support and enthusiasm for outdoor
EE, and a composite score can be used as a control variable to test whether treatment differences
maintain their effect when teacher or school effects are constant. However, sample sizes must be large
enough for researchers to compare subgroups (Rosenbaum, 1991; Rosenbaum & Rubin, 1983).
Mail surveys suffer from nonresponse bias. Case studies of gardens are often first-person accounts
by their initiators. In each case of bias, results may appear more positive than they are in reality.
Qualitative results are applicable only to the situation studied and can be used to form a hypothesis,
but researchers cannot extend them to other situations. However, there is consistency in the results of
the seven reviewed qualitative studies in Tables 5 and 6 from around the United States, which would
lead researchers to accept the legitimacy of their findings.
Overall, the methodology in the evaluation of school-gardening programs needs to be much more
rigorous. Ozer (2007) suggested a combination of systematic qualitative and quantitative methods
emphasizing direct observation because implementation of school gardening cannot be assumed to
be uniform, even within the same school. Phibbs and Relf (2005) suggested longitudinal research.
Students should act as their own controls, as in McAleese and Rankin’s (2007) study. Outcomes
should be documented in ways that will affect educational policy toward school gardening and
subsequent funding.

Environmental and Social Impacts of School Gardens

From an environmental perspective, school gardens may seem to researchers to be a limited sub-
stitute for redesigning the whole schoolyard or for interacting more closely with nearby landscapes.
However, Moore (1995) reported that the school vegetable garden was the most feasible pedagogical
vehicle for promoting daily environmental learning in his project’s fully redesigned schoolyard. In
particular, annual vegetable and flower gardens enabled a yearly full start from bare soil. Each year,
students could be full participants in designing the garden and the act of regeneration, the regenera-
tive act of embedding tiny seeds in dirt and food-scrap compost, and nurturing those seeds during
their transformation into flowers or vegetables. Researchers have frequently commented on how
excited children were to put their hands in dirt. Birds, insects, spiders, weeds, and mammalian preda-
tors were players in this process, so that the school’s environmental complexity flourished. Food pro-
duction connected students to sensual pleasures, sustenance, and the agri-systems of daily life in and

34 THE JOURNAL OF ENVIRONMENTAL EDUCATION

beyond the school. Butterfly, habitat, and pond gardens required less summer maintenance and also
provided diverse opportunities for observing natural systems. Educators found that gardens produced
the ecological diversity that had been missing from monocultural schoolyards, allowing children to
work directly with energy transformation and entropy. A school garden can be a frequent, if not daily,
experience, a place owned by the students. With attention, gardens can also create delight and pride
and foster the kind of unfettered play that children create in simple hidden spaces (Mergen, 2003).
These gardens can be private spots for observing, fostering the imagination, or simply reading among
the fava beans (Brynjegard, 2001). Uniformly, the qualitative studies of kindergarten to sixth-grade
gardening that are summarized in Tables 5 and 6 showed the following positive behavioral and social
outcomes: heightened motivation and enthusiasm, improved sense of self, teamwork, community,
and parental involvement. The amount of structure that children encounter in relating to a garden
may determine whether they benefit both cognitively and emotionally (Thorp & Townsend, 2001).
How much structure is appropriate is an area for further school-gardening research.

Teachers’ Need for Support and Training

The teacher and principal are major variables in school-garden success. In particular, more needs
to be known about the principal’s effect. Major teacher issues are lack of personal interest and limited
capabilities, knowledge, and time. Samples of gardening teachers and teachers with adequate garden-
ing support (Graham et al., 2004) were more enthusiastic about the potential of school gardens than
were mixed samples of gardening and nongardening teachers (Graham & Zidenberg-Cherr, 2005).
Support may come in the form of (a) enthusiastic principals, (b) effective and credible lead teachers
who promote school gardening through contagious student excitement rather than through personal
power (Vesilind & Jones, 1998), or (c) the semiretirement-lead gardener programs for teachers, such
as at the Davis, California, schools (Graham et al.). In Las Vegas, sequential surveys of principals in
gardening and nongardening schools regarding potential problems and barriers to school gardening
led to the hiring of a community-based instructor to provide training and coordinate the gardening
program and the volunteer master gardeners (O’Callaghan, 2005).
Additional studies are necessary on how educators can best remove barriers to implementing and
keeping school gardens running. Studies have not addressed school-garden continuity or failure, but
they have addressed the lack of teacher preparation for using gardens in instruction. Portillo (2002)
reported that elementary teachers with some agricultural training are more likely to use school gardens
as a learning tool. Dobbs, Relf, and McDaniel (1998) reported that 98% of the 205 Virginia kindergar-
ten to sixth-grade teachers whom they surveyed wanted to participate in additional gardening training.
School-gardening experience and plant science could become a part of teachers’ preservice education, so
that all teachers can feel prepared to use school gardening as a potent form of experiential education.

Conclusions and Suggestions for Further Research

The results of qualitative, quantitative, and survey research have supported the conclusion that
school gardening can improve students’ test scores and school behavior. Teachers believe that gardens
promote academic instruction. However, methodological shortcomings of the quantitative studies
have reduced faith in these results. Gardens can improve the ecological complexity of the schoolyard
in ways that promote effective experiential learning in many subject areas, particularly the areas
of science, EE, and food education. Researchers and educators should pay attention to how they
design the garden and the learning experience in the garden. Both preservice and in-service teach-
ers need more training to effectively use gardening as a teaching tool. Teachers are the mainstay
of school gardening. However, gardens require embedded support mechanisms that lighten the
teacher’s burden.

WINTER 2009, VOL. 40, NO. 2 35

 To improve school-gardening outcomes research, researchers need to (a) use well-designed longitu-
dinal studies that combine qualitative and quantitative elements with appropriate sample design, (b)
use validated instruments, and (c) control for teacher and ethnicity effects. More qualitative studies
of smoothly functioning school gardens that examine how success is managed and maintained are
also necessary. Other productive future research would be (a) studies of reasons for garden failure
and (b) reports on creative means of maintaining gardens over time and moving the workload away
from teachers. More research would be useful on the level of structure versus self-exploration in a
garden that best serves the student’s learning needs. Researchers and educators also need to know
whether the changes in environmental sensitivity and observation skills reported in qualitative studies
of gardening are transient or long lasting, affecting behavior as a child matures.

REFERENCES
Alexander, J., North, M.-W., & Hendren, D. K. (1995). Master gardener classroom garden project: An evaluation of the
benefits to children. Child Environments, 12(2), 124–133.
Blair, D. (1996). Eating in the bioregion. In J. Chesworth (Ed.), The ecology of health: Identifying issues and alternatives (pp.
297–307). Thousand Oaks, CA: Sage.
Blair, D., Giesecke, C. C., & Sherman S. (1991). A dietary, social and economic evaluation of the Philadelphia urban
gardening project. Journal of Nutrition Education, 23, 161–167.
Blair, D., & Sobal, J. (2006). Luxus consumption: Wasting food resources through overeating. Agriculture and Human
Values, 23(1), 63–74.
Bloom, B. S., Engelhart, M. B., Furst, E. J., Hill, W. H., & Krathwohl, D. R. (1956). Taxonomy of educational objectives:
Handbook I. The classification of educational goals–Cognitive domain. New York: Longman.
Brink, L., & Yoast, B. (2004). Transforming inner-city school grounds: Lessons from learning landscapes. Children, Youth
and Environments, 12, 208–232.
Brunotts, C. M. (1998). School gardening: A multifaceted learning tool. An evaluation of the Pittsburgh civic garden center’s
Neighbors and Schools Gardening Together. Unpublished masters thesis, Duquesne University, Pittsburgh, PA.
Brynjegard, S. (2001). School gardens: Raising environmental awareness in children. San Rafael, CA: School of Education,
Dominican University of California. (ERIC Documentation Reprroduction Service No. ED452085). Retrieved
December 23, 2006, from http://edres.org/eric/ED452085.htm
Bundschu-Mooney, E. (2003). School garden investigation: Environmental awareness and education. San Rafael, CA: Division
of Education, School of Business, Education and Leadership, Dominican University of California. (ERIC Document
Reproduction Service No. ED480981). Retrieved August 2, 2006, from http://edres.org/eric/ED480981.htm
California Department of Education. (2005). Catalogue listing of publications. Retrieved July 8, 2006, from
http://www.cde.ca.gov/re/pn/rc/ap/pubdisplay.aspx?ID=001512
Canaris, I. (1995). Growing foods for growing minds: Integrating gardening into the total curriculum. Children’s
Environments, 12(2), 134–142.
Chawla, L. (1998). Significant life experiences revisited: A review of research on sources of environmental sensitivity. The
Journal of Environmental Education, 29(3), 11–21.
Corson, C. (2003). Grounds for learning: Hope for America’s derelict schoolyards. Retrieved August 23, 2008, from http://
www.cherylcorson.com/publications.html
Culin, J. D. (2002). Butterflies are great teachers: The South Carolina butterfly project. American Entomologist, 48(1),
14–18.
DeMarco, L. W., Relf, D., & McDaniel, A. (1999). Integrating gardening into the elementary school curriculum.
HortTechnology, 9, 276–281.
Demas, S. (1979). School gardens and environmental education. Nature Study, 32(3), 3–5.
Dirks, A. E., & Orvis, K. (2005). An evaluation of the junior master gardener program in third grade classrooms.
HortTechnology, 15, 443–447.
Dobbs, K., Relf, D., & McDaniel, A. (1998). Survey on the needs of elementary education teachers to enhance the use of
horticulture or gardening in the classroom. HortTechnology, 8, 370–373.
Donald, M. N. (1960). Implications of nonresponse for the interpretation of mail questionnaire data. Public Opinion
Quarterly, 24(1), 99–114.
Ellis, R. A., Endo, C. M., & Armer, J. M. (1970). The use of potential nonrespondents for studying nonresponse bias.
Pacific Sociological Review, 13(2), 103–109.
Emekauwa, E. (2004). They remember what they touch: The impact of place-based learning in East Feliciana parish. Rural
School and Community Trust, Arlington, VA. Retrieved June 3, 2006, from http://www.peecworks.org/PEEC/PEEC_
Research/S0009D4FB-0084FE88
Faddegon, P. A. (2005). The kids growing food school gardening program: Agricultural literacy and other educational outcomes.
Doctoral dissertation, Cornell University, Ithaca, NY.

36 THE JOURNAL OF ENVIRONMENTAL EDUCATION

Finch, K. (2004). Extinction of experience: A challenge to nature centers? (Or, how do you make a conservationist?)
Directions: The Journal of the Association of Nature Center Administrators, Special Issue, 1–7.
Francis, M. (1995). Childhood’s garden: Memory and meaning of gardens. Children’s Environments, 12(2), 1–16.
Gaster, S. (1991). Urban children’s access to their neighborhood changes over three generations. Environment and Behavior,
23(1), 70–85.
Gibbs, W. W., & Fox, D. (1999). The false crisis in science education. Scientific American, 281(10), 87–93.
Graham, H., Beall, D. L., Lussier, M., McLaughlin, P., & Zidenberg-Cherr, S. (2005). Use of school gardens in academic
instruction. Journal of Nutrition Education Behavior, 37(3), 147–151.
Graham, H., Feenstra, G., Evans, A. M., & Zidenberg-Cherr, S. (2004). Davis school program supports life-long healthy
eating habits in children. California Agriculture, 58, 200–205.
Graham, H., & Zidenberg-Cherr, S. (2005). California teachers perceive school gardens as an effective nutritional tool to
promote healthful eating habits. Journal of American Dietetic Association, 105, 1797–1800.
Hudkins, S. J. (1995). Parvis e glandibus quercus: “Great oaks from little acorns grow.” Journal of Extension, 33(4).
Retrieved September 20, 2006, from http://www.joe.org/joe/1995august/iw6.html
Kahn, P. H., Jr. (2002). Children’s affiliations with nature: Structure, development and the problem of environmental gen-
erational amnesia. In P. H. Kahn Jr. & S. R. Kellert (Eds.), Children and nature: Psychology, sociocultural and evolutionary
investigations (pp. 94–116). Cambridge, MA: MIT Press.
Kellert, S. R. (2002). Experiencing nature: Affective, cognitive, and evaluative development in children. In P. H. Kahn
Jr. & S. R. Kellert (Eds.), Children and nature: Psychology, sociocultural and evolutionary investigations (pp. 117–151).
Cambridge, MA: MIT Press.
Klemmer, C. D., Waliczek, T. M., & Zajicek, J. M. (2005a). Development of a science achievement evaluation instrument
for a school garden program. HortTechnology, 15, 433–438.
Klemmer, C. D., Waliczek, T. M., & Zajicek, J. M. (2005b). The effect of a school gardening program on the science
achievement of elementary students. HortTechnology, 15, 448–552.
Krasny, M., & Doyle, R. (2002). Participatory approaches to program development and engaging youth in research: The
case of an inter-generational urban community gardening program. Journal of Extension, 40(5), 1–21.
Lekies, K., Eames-Sheavly, M., Wong, K., & Ceccarini, A. (2006). Children’s garden consultants: New model of engaging
youth to inform garden design and programming. HortTechnology, 16, 139–142.
Lieberman, G. A., & Hoody, L. L. (1998). Closing the achievement gap: Using the environment as an integrating context for
learning. Paper presented at the State Education and Environmental Roundtable, San Diego, CA.
LifeLab. (2006). Garden based learning that promotes a sustainable future. Retrieved July 31, 2006, from http://www.lifelab
.org/about/index.html
Lineberger, S. E., & Zajicek, J. M. (2000). School gardens: Can a hands-on teaching tool affect students’ attitudes and
behaviors regarding fruits and vegetables? HortTechnology, 10, 593–597.
Lohr, V., & Person-Mims, C. H. (2005). Children’s active and passive interactions with plants influence their attitudes and
actions as adults. HortTechnology, 15, 472–476.
Mabie, R., & Baker, M. (1996). The influence of experiential instruction on urban elementary students’ knowledge of the
food and fiber system. Journal of Extension, 34(6), 1–4.
McAleese, J. D., & Rankin, L. L. (2007). Garden based nutrition education affects fruit and vegetable consumption in
sixth grade adolescents. American Dietetic Association, 107, 662–665.
Mergen, B. (2003). Review essay: Children and nature in history. Environmental History, 8, 643–669.
Milton, B., Cleveland, E., & Bennet-Gates, D. (1995). Changing perceptions of nature, self, and others: A report on a
park/school program. The Journal of Environmental Education, 26(3), 32–39.
Moore, R. (1995). Growing foods for growing minds: Integrating gardening and nutrition education into the total cur-
riculum. Children’s Environments, 12(2), 134–142.
Morris, J., Briggs, M., & Zidenberg-Cherr, S. (2000). School-based gardens can teach kids healthier eating habits.
California Agriculture, 54(5), 40–46.
Morris, J. L., Neustadter, A., & Zidenberg-Cherr, S. (2001). First-grade gardeners more likely to taste vegetables. California
Agriculture, 55(1), 43–46.
Morris, J. L., & Zidenberg-Cherr, S. (2002). Garden-enhanced nutrition education curriculum improves fourth-grade
school children’s knowledge of nutrition and preferences for some vegetables. Journal of American Dietetic Association,
102(1), 91–93.
Nabhan, G. P., & Trimble, S. (1994). The geography of childhood: Why children need wild spaces. Boston: Beacon Press.
National Environmental Education & Training Foundation. (2000). Environmental-based education: Creating high perfor-
mance schools and students. Retrieved May 3, 2006, from http://www.neetf.org/pubs/index.htm
National Gardening Association. (2006). About NGA. Retrieved August 1, 2006, from http://assoc.garden.org/about
O’Callaghan, A. M. (2005). Creating a school gardens program in the challenging environment of Las Vegas, Nevada.
HortTechnology, 15, 429–433.
Orr, D. W. (2002). Political economy and the ecology of childhood. In P. H. Kahn Jr. & S. R. Kellert (Eds.), Children and
nature: Psychology, sociocultural and evolutionary investigations (pp. 279–303). Cambridge, MA: MIT Press.
Ozer, E. J. (2007). The effects of school gardens on students and schools: Conceptualization and considerations for maxi-
mizing healthy development. Health Education and Behavior, 34, 846–863.

WINTER 2009, VOL. 40, NO. 2 37

 Phibbs, E. J., & Relf, D. (2005). Improving research on youth gardening. HortTechnology, 15, 425–428.
Pollan, M. (2006). Omnivore’s dilemma: A natural history of four meals. New York: Penguin Press.
Portillo, M. T. (2002). An assessment of agricultural knowledge of kindergarten through sixth grade teachers. Unpublished
doctoral dissertation, Oklahoma State University, Stillwater, OK.
Rahm, J. (2002). Emergent learning opportunities in an inner-city youth gardening program. Journal of Research in Science
Teaching, 39, 164–184.
Robinson, C. W., & Zajicek, J. M. (2005). Growing minds: The effects of a one-year school garden program on six con-
structs of life skills of elementary school children. HortTechnology, 15, 453–457.
Rosenbaum, P. R. (1991). Discussing hidden bias in observational studies. Annals of Internal Medicine, 115, 901–905.
Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal
effects. Biometrika, 70(1), 41–55.
Skelly, S. M., & Bradley, J. C. (2000). The importance of school gardens as perceived by Florida elementary school teach-
ers. HortTechnology, 10, 229–231.
Skelly, S. M., & Zajicek, J. M. (1998). The effect of an interdisciplinary garden program on the environmental attitudes
of elementary students. HortTechnology, 8, 579–583.
Smith, L. L., & Mostenbocker, C. E. (2005). Impact of hands-on science through school gardening in Louisiana public
elementary schools. HortTechnology, 15, 439–443.
Subramaniam, A. (2002, Summer). Garden-based learning in basic education: A historical review. Monograph, 1–11.
Retrieved June 3, 2006, from http://fourhcyd.ucdavis.edu/publications/monograph.html
Sullivan, A. F. (1999). Community gardening in rural regions: Enhancing food security and nutrition. Retrieved June 1, 2006,
from Tufts University Center on Hunger and Poverty Web site: http://nutrition.tufts.edu/
Thorp, L., & Townsend, C. (2001, December 12). Agricultural education in an elementary school: An ethnographic study
of a school garden. Proceedings of the 28th Annual National Agricultural Education Research Conference in New Orleans,
LA (pp. 347–360). Retrieved from http://www.aaaeonline.org/conference_files/758901
University of Florida. (2006). Florida school garden competition. Retrieved July 21, 2006, from http://hort.ufl.edu/ggk/
comp.htm
U.S. Department of Agriculture. (2006). Economic Research Service. Briefing rooms: Rural population and migration.
Retrieved December 23, 2006, from http://www.ers.usda.gov/briefing/Population/
Vaske, J. J., & Kobrin, C. K. (2001). Place attachment and environmental responsible behavior. The Journal of
Environmental Education, 32(4), 16–21.
Vesilind, E. M., & Jones, M. C. (1998). Gardens or graveyards? Science education reform and school culture. Journal of
Research in Science Teaching, 35, 757–775.
Waliczek, T. M., Bradley, R. D., Lineberger, R. D., & Zajicek, J. M. (2000). Using a web-based survey to research the
benefits of children’s gardening. HortTechnology, 10, 71–76.
Waliczek, T. M., Bradley, R. D., & Zajicek, J. M. (2001). The effect of school gardens on children’s interpersonal relation-
ships and attitudes toward school. HortTechnology, 11, 466–468.
Waliczek, T. M., Logan, P., & Zajicek, J. M. (2003). Exploring the impact of outdoor environmental activities on children
using a qualitative text data analysis system. HortTechnology, 13, 684–688.
Waliczek, T. M., & Zajicek, J. M. (1999). School gardening: Improving environmental attitudes of children through
hands-on learning. Journal of Environmental Horticulture, 17, 180–184.

38 THE JOURNAL OF ENVIRONMENTAL EDUCATION